This comprehensive high-level design document outlines a scalable food delivery platform architecture similar to Zomato and Swiggy. The system supports 50M+ users and 5M daily orders through a microservices architecture with polyglot persistence. Key components include: REST APIs for core features (user management, restaurant discovery, cart, orders), WebSocket servers for real-time tracking, and message queues (Kafka) for asynchronous communication.

The data layer uses PostgreSQL for transactional data (orders, payments), MongoDB for restaurant/menus, Elasticsearch for search, Redis for caching/sessions, and Cassandra for location tracking. The design emphasizes 99.9% availability, <500ms restaurant listing latency, and horizontal scalability. Critical flows include order placement with payment integration, delivery partner assignment, and live tracking with ETA calculation.

Performance optimizations include multi-layer caching (CDN, Redis), database indexing, connection pooling, and rate limiting. Edge case handling covers payment failures, no delivery partners, item unavailability, and concurrent order prevention using distributed locks.